College aantekeningen
Quantitative Methods MAN-BCU2030EN all summary of necessary literature
Quantitative Methods MAN-BCU2030EN all summary of necessary literature
[Meer zien]Voorbeeld 4 van de 90 pagina's
Voorbeeld 4 van de 90 pagina's
In winkelwagengesponsord bericht van onze partner
Voorbeeld van de inhoud
ContentsLecture 1 – Linear regression .................................................................................................................. 2
Lineaire Regressie (stuvia) ....................................................................................................................... 5
Lecture 2 – Discrete choice models....................................................................................................... 11
Discrete keuzemodellen (Stuvia) ........................................................................................................... 17
Logistische regressie/binaire keuze modellen (binominaal meetniveau) (Stuvia) ........................... 17
Multiple logistische regressie (Stuvia) .............................................................................................. 20
Multinomiale keuze modellen (Stuvia) ............................................................................................. 21
Lecture 3 – Paper-lecture – discrete choice models ............................................................................. 23
Lecture 4 – Spatial analysis.................................................................................................................... 25
Lecture 5 – Temporal dimension and quantitative analysis ................................................................. 35
Temporale dimensie: tijdreeksanalyse en forecasting (Stuvia) ............................................................ 42
Lecture 6 – Paper-lecture – temporal dimensions and quantitative analysis....................................... 46
Lecture 7 – Structural equation modelling, introduction, concepts, assumptions and causal models 48
Lecture 8 – Structural equation modelling, identification, evaluation, measurement models ............ 55
Lecture 9 – Structural equation modelling, measurement scale, hybrid models ................................. 62
Structuurvergelijkingsmethoden (Stuvia) ............................................................................................. 69
Lecture 10 – Structural equation modelling – Paper lecture ................................................................ 76
Lecture 11 – Latent class analysis.......................................................................................................... 79
Latente klasse analyse (Stuvia).............................................................................................................. 87
Ieder model apart toegelicht................................................................................................................. 89
1
, Stuvia samenvatting kwantitatieve
methoden & aantekeningen colleges
Lecture 1 – Linear regression
Quantitative methods is basically a top down and deductive (from a general law) approach. It is
about relationships between variables. One variable is the cause and the other variable the
consequence (impact, effect). (i.e. weather have an effect on travel behavior).
Start with a theory, then deriving hypothesis, empirical reality (data collection), findings,
interpretation.
Point of attention in data collection:
• Reliability: internal and external validity
o Internally valid: are the questions questioning the subject you want to question?
o External valid: The power of quantitative is large scale. You present the survey to a
group of representatives. Your sample population needs to have the right
characteristics, so it is applicable to the whole population you want to examine.
• Operationalization of theories and concepts
o Concepts/theories → dimensions → indicators
• Questionnaire design: measurement and scale construction.
• Interpretation: correlation vs causality & synthesis (what can you conclude at a higher
level?) and generalization.
o Statistics doesn’t tell you about causality, as a researcher you need to do that, the
cause and the consequence!
There are different methods of data collection for quantitative methods:
• Experiment and randomization
• Survey/questionnaire (population vs sample / a select (randomly chosen, no selective bias,
generally assume earlier that it realize the characteristics of the population) vs non-select
• Longitudinal research → timeseries (each individual data of three monthly dates, so you see
individual improvements etc.) vs panel data research (data that is derived from a (usually
small) number of observations over time on a (usually large) number of cross-sectional units
like individuals, households, firms, or governments)
A sample is random so, the larger the population, the better it represents the characteristics of this
population.
2
,Variable types and methods of analysis:
Response variable also dependent variable (= phenomena that you want to explain)
Explanatory variable also independent variable (= phenomena aspects that are explaining this
outcome, i.e. age).
Manifest variable = observed variables
Latent variable = not directly measurable (IQ via math, etc.) An unobserved variable.
Quantitative = discrete (you score one in soccer goal, you can’t score 1/2e goal) and continuous any
number with .25 or .33. It is endless.
The measurement level determines the possibilities we have. Interested in the measurement level op
de independent variable.
Nominal scale → grouping, categories that are not ordered
Binominal → yes, no. male, female.
Ordinal scale → Order them in meaningful way → low, medium, high
Interval scale → equal distances between categories. Age 0-10, 11-20 etc.
Ratio scale → 0 has a meaningful value
Metric → sales, costs, number of clients etc.
Non-metric → gender, nationality etc.
Linear regression → with metric and continuous variables. Hight, weight, housing prices.
Linear regression → you have a certain outcome, numeric, you want to explain with some other
features i.e. the garbage consumption of municipalities (dependent variable), want to explain that
variable based on different features like: number of inhabitants, recycling habit, geographical
location, size of the city. Kortom: How does the explanatory variable (independent) influencing the
explaining variable (dependent).
Y= observations of number of municipalities
X = explanatory variables (number of inhabitants, recycling habit etc.)
Linear model, so we can add them up (+)
B0= intercept, this is were it start, we all x are 0, we still have a Y value.
Specific observation deviates randomly from the expected value, so we add a random error
We want to know how the explanatory variables influence the outcome. If we add one household
(explanatory variable) how much will the garbage production change.
Slope = response in garbage change by households.
Model = never true reality. Specific observation deviates randomly from the expected value, so we
add a random error.
3
, OLS = Ordinary Least Square = which model is the closest to reality.
Residual “e” : Difference between the observed and the predicted value. Basically, an error. When
the residual is calculated for each observation in a data set the resulting collection as referred to as
the residuals. The higher the correlation, the lower the residual variance around the line will be.
Find an optimal linear line which is closest to reality.
318 dots, which share of total variation of the garbage production is explained by this model → this
is R2 actually doing. R2 need to between 0% and 100%. By 0% = no explanatory power of our
model. 0.1 = zeer zwak, 0.9 = zeer sterk.
Linear regression model: R2 is looking at the share of explained variation out of the total variation.
R2 = 100% meaning → every dot is on the linear line. The smaller the R2, how further the dots are.
Model assumptions if it satisfies the conditions for linear regression→ if faith in the model
• Observations are independent → you don’t want clustering of the datapoints. Can not be
tested in the model, but research design.
• A linear model is suitable, check the plotting. So, relation between garbage production and
households. You do not want a pattern, but a linear relationship. You want to have a random
distribution.
• We want to have a constant variance along the range. The variance of the residuals is equal
for all possible values of the independent variables. We want to have a parameter which is
reliable and want the reliability across the range.
• The residuals are normally distributed→ histogram
Outliers (observations which are entirely different than the average
observations) can be a problem for the whole linear line and the
general outcome. We can detect them:
Look at observations beyond 3 standard deviation of the mean and
via boxplots, histograms, probability plots, scatterplots. But you can
also see them in the model itself.
Multicollinearity issue: the correlation between two (or more) explanatory variables is too high
(r<0.8 or 0.9) and we can’t distinguish the different effects. (households and individuals, if the
correlation is too high we get problems because we can’t distinguish different effects of the variables
of each other). You can then exclude one of your variables from the analysis.
Creating dummy variables → variables which is having two levels. 0 and 1. Man = 1, women = 0
Distance between two lines is constant throughout the range. Why is this useful? → if we do this, we
see a more logical pattern if we add this variable.
4
Voordelen van het kopen van samenvattingen bij Stuvia op een rij:
Verzekerd van kwaliteit door reviews
Stuvia-klanten hebben meer dan 700.000 samenvattingen beoordeeld. Zo weet je zeker dat je de beste documenten koopt!
Snel en makkelijk kopen
Je betaalt supersnel en eenmalig met iDeal, creditcard of Stuvia-tegoed voor de samenvatting. Zonder lidmaatschap.
Focus op de essentie
Samenvattingen worden geschreven voor en door anderen. Daarom zijn de samenvattingen altijd betrouwbaar en actueel. Zo kom je snel tot de kern!
Veelgestelde vragen
Wat krijg ik als ik dit document koop?
Je krijgt een PDF, die direct beschikbaar is na je aankoop. Het gekochte document is altijd, overal en oneindig toegankelijk via je profiel.
Tevredenheidsgarantie: hoe werkt dat?
Onze tevredenheidsgarantie zorgt ervoor dat je altijd een studiedocument vindt dat goed bij je past. Je vult een formulier in en onze klantenservice regelt de rest.
Van wie koop ik deze samenvatting?
Stuvia is een marktplaats, je koop dit document dus niet van ons, maar van verkoper koendeweert. Stuvia faciliteert de betaling aan de verkoper.
Zit ik meteen vast aan een abonnement?
Nee, je koopt alleen deze samenvatting voor €10,99. Je zit daarna nergens aan vast.
Is Stuvia te vertrouwen?
4,6 sterren op Google & Trustpilot (+1000 reviews)
Afgelopen 30 dagen zijn er 80796 samenvattingen verkocht
Opgericht in 2010, al 14 jaar dé plek om samenvattingen te kopen